Lida Ungar

Learn More
In this paper, we present a set of techniques for the evaluation of brain tissue classifiers on a large data set of MR images of the head. Due to the difficulty of establishing a gold standard for this type of data, we focus our attention on methods which do not require a ground truth, but instead rely on a common agreement principle. Three different(More)
Disturbances in selective attention represent a core characteristic of schizophrenia, whose neural underpinnings have yet to be fully elucidated. Consequently, we recorded brain activation using functional magnetic resonance imaging (fMRI) while 15 patients with schizophrenia and 15 age-matched controls performed a well-established measure of selective(More)
In this paper, we present an evaluation of seven automatic brain tissue classifiers based on level of agreements. A number of agreement measures are explained, and we show how they can be used to compare different segmentation techniques. We use the Simultaneous Truth and Performance Level Estimation (STAPLE) of Warfield et al. but also introduce a novel(More)
We present a quantitative evaluation of MR brain images segmentation. Five classifiers were tested. The task was to classify an MR image into four different classes: background, cortical spinal fluid, gray matter and white matter. The performance was rated by first estimating a ground truth (EGT) using STAPLE and then analyzing the volume differences as(More)
Attention and memory deficits are among the most prominent cognitive disturbances observed in schizophrenia. It has been suggested that a disruption in anatomical connectivity between areas involved in attentional control might be responsible for these abnormalities. We used Diffusion Tensor Tractography and Color Stroop/Negative Priming(NP) paradigm to(More)
Objective: We are interested in finding a way of comparing and evaluating different segmentation techniques in a robust and automatic way. In this project, we have compared five different brain tissue classifiers applied on 40 different MR images. The evaluation is based on the notion of agreement, i.e. how consistent is a specific technique with all the(More)
  • 1